Hydroelectric generating set generator fault diagnosis method based on efficiency and loss

Abstract

The invention provides a hydroelectric generating set generator fault diagnosis method based on efficiency and loss, and the method comprises the steps: firstly introducing energy and subentry loss parameters to an energy conversion device, i.e., a generator, and enabling the parameters to serve as state evaluation and fault diagnosis features; designing a health state model based on energy efficiency and sub-item loss to utilize the characteristics; and designing a generator operation state evaluation and diagnosis method based on the energy and loss combination characteristics, namely, energy state evaluation is carried out by adopting efficiency parameters, sub-item loss characteristics of the current working condition are obtained through sub-item loss state evaluation according to theevaluation result, and finally, fault diagnosis is carried out to obtain the detailed faults of the generator. The energy efficiency characteristic is used as the generator state evaluation characteristic, and the operation state of the unit can be effectively reflected. According to the method, performance evaluation can still be achieved for the generator set which does not frequently have sudden faults, and a basis is provided for life estimation and trend prediction.

Description

technical field [0001] The invention relates to the technical field of hydroelectric generating sets, in particular to a method for diagnosing faults of hydroelectric generating sets based on efficiency and loss. Background technique [0002] For conventional hydroelectric generator sets, there are a large number of physical quantity sensors to monitor its operating status in real time, such as the swing of the guide bearing, the temperature of the bearing bush, the flow rate of the cooling medium, the temperature of the cooling medium, the speed of the unit, etc.; at the same time , A large number of electrical parameters are also sent to the monitoring system in real time, such as excitation voltage, excitation current, machine terminal voltage, machine terminal current, active power, power factor and so on. [0003] Therefore, for possible sudden failures of generators, such as stator winding grounding or severe vibration at the machine end, protection systems and monitor...

AI Technical Summary

Problems solved by technology

The problem is that it is mainly based on monitoring and supplemented by alarm; in essence, it is still a traditional condition monitoring system, that is, to obtain various characteristic quantities of the unit for display and comparison of basic limit values, which belongs to the alarm system, not the real diagnosis. system

The problem is: the normal faults of the generator are basically managed and handled by the protection and monitoring system. The various fault rules recorded in the expert knowledge base are usually only valid for instantaneous or sudden faults, and a large number of insulation and aging of the generator may cause potential faults, but cannot be matched and located in real time through rules

The problem is that a single parameter such as vibration, swing or pressure cannot show the performance degradation trend of the unit, but can only reflect the physical state of a certain part or a certain position of the unit. Some schemes will consider using a combination of characteristic parameters, such as several The ratio or correlation of physical quantities, but there is no method that uses energy and efficiency as diagnostic features.

In addition, taking the large shaft swing of the unit as an example, it is affected by the clearance of the bearing pad. When the bearing pad is fastened by turning, the large shaft swing will naturally decrease, but this will increase the friction of the bearing pad. and the increase of the tile temperature, therefore, it is difficult to judge the operating characteristics of the unit through the isolated physical quantity characteristics

In addition, this method cannot obtain the specific faults of the final unit, and can only achieve the abnormal state evaluation results of simple parameters

Images